Solving the Speech Recognition Accent Gap with Global English in Speech-to-Text: A Case Study of Ghana

Abstract

In recent years, Speech recognition has become the state of the art for speech to text programs. However, speech recognition systems often struggle to correctly interpret foreign accents or non-native English. We aim to contribute to the domain of literature on speech to text by finding ways of improving the performance of speech recognition whilst dealing with Ghanaian accented English speech by increasing non-standard English audio samples and the training dataset of Automatic Speech Recognition